AI Credits
Last updated
Was this helpful?
Last updated
Was this helpful?
Every plan in Relay.app comes with a generous number of free AI credits. AI credits are a single pool of credits that are consumed when running steps that use AI-related features, including:
Prompting AI models
Audio transcription and text-to-speech
Website scraping
The free AI credits that come with all our subscriptions are good for reasonable usage—see the tables below for more details and examples.
If you run out of AI credits in Relay.app, purchase more in your , or contact support@relay.app for help. Alternatively, if you already pay for credits with a AI service provider, connect that account to Relay.app and switch your AI automations to use that connected model.
Free
Professional
Team
When the AI credits included in your plan aren't enough, it's easy to add more via the . Note that you need to be on a paid plan in order to be able to add additional AI credits. If you're on a Free plan, upgrade to Professional or Team first.
It depends! Different models and operations require different numbers of credits. A detailed conversion table is below, but here some concrete examples:
Summarizing an email will typically uses about 1 AI Credit (but may be much less)
Scraping text from a website uses about 5 AI Credits
Transcribing a minute of audio uses about 10 AI Credits
Updated: 2025-03-27
Web scraping: general
-> 5 AI credits per website scraped
Web scraping: Google Search
-> 5 AI credits per Google Search performed
Web scraping: Amazon, Zillow, Indeed
-> 1.67 AI credits per number of Profiles/Pages requested
Web scraping: LinkedIn
-> 7 AI credits per request
Web scraping: YouTube video transcript
-> 16 AI credits per transcript
Web scraping: YouTube video metadata
-> 1.67 AI credits per metadata request
OpenAI Audio Transcription
6s of transcribed audio
OpenAI Text-to-Speech (TTS)
Fast: 40 characters converted to speech High definition: 20 characters converted to speech
DALL-E 3 image generation
-> 140 AI credits per image
Anthropic Claude 3 Haiku
Anthropic Claude 3.5 Haiku
Anthropic Claude 3 Opus
Anthropic Claude 3.5 Sonnet
Anthropic Claude 3.7 Sonnet
Google Gemini 1.5 Flash
<128k context:
Google Gemini 2.0 Flash Lite Preview
Google Gemini 2.0 Flash
Google Gemini 1.5 Pro
<128k context:
OpenAI GPT-4o mini
OpenAI GPT-4o
OpenAI o1
OpenAI o3-mini
Perplexity Sonar Reasoning
Perplexity Sonar
Perplexity Sonar Pro
DeepSeek V3
DeepSeek R1
The definition of a token differs slightly by model provider. Additionally, input and output tokens are also differentiated from a cost/consumption perspective. We suggest reading the short FAQs from each model provider to learn more:
Unfortunately, it's not trivial to predict how many tokens an AI step will consume upfront, although the articles linked above will give you a better understanding.
For AI steps in Relay.app, token usage is primarily influenced by these factors:
Context included in the prompt The more and larger the attached context, the more tokens will be used every time your step is executed. For example:
An email can easily increase token usage by a few thousand
A PDF can go way beyond that, depending on the size
Whether the step is given Internet access All text on the websites that the model needs to visit for your prompt is included as context, including user-hidden HTML. This can be anywhere between a few hundred to tens of thousands of tokens used per step execution, depending on how many websites the model needs to visit and how large they are.
480 output
2400 input
120 output
600 input
8 output
40 input
40 output
200 input
40 output
200 input
2k output
8k input >128k context: 1k output 4k input
2k output 8k input
1.5k output
6k input
120 output
480 input >128k context: 60 output
240 input
1k output 4k input
60 output 240 input
10 output 40 input
50 output 200 input
120 output 600 input
600 output 600 input
40 output 200 input
500 output 2k input
270 output 1090 input